generated: '2026-07-20'
method: searched
source: https://github.com/NebuSec/vega-skill/blob/main/skills/vega-cli/SKILL.md
note: >-
  Cross-cutting request/response semantics for the Vega backend, captured from the
  published Vega CLI reference (no public OpenAPI). Cross-links authentication/,
  errors/, and data-model/.
authentication:
  style: bearer-api-key-or-oauth
  ref: authentication/nebula-security-authentication.yml
identifiers:
  style: self-describing-prefixed
  prefixes:
    project: pg_
    repository: proj_
    scan: scan_
    finding: finding_
  note: Wherever a <project> or <repo> argument is accepted, a unique name also resolves.
pagination:
  style: cursor
  request_param: --cursor
  response_field: next_cursor
  fetch_all: --all
  note: In text mode the cursor prints on stderr; in --json mode next_cursor is in the body.
output_modes:
  default: aligned columns (lists) / markdown-style sections (details) on stdout
  json: global --json flag returns raw backend JSON (compact, one object/array per line / NDJSON for multi)
  stream_channels: stdout=data only; stderr=progress, warnings, errors
request_tracing:
  id_field: request_id
  surfaced_in: "error messages: error[<code>]: <message> (request_id=…)"
error_envelope:
  ref: errors/nebula-security-error-codes.yml
  cli_format: "error[<code>]: <message> (request_id=…)"
filtering:
  params: [--severity, --status, --file-prefix, --cwe, -q, --limit]
idempotency:
  supported: false
  note: No idempotency-key contract is documented for the Vega backend/CLI.
cost_governance:
  note: Scan-creating operations require explicit cost consent (--max-cost / --yes) and support a server-side spend cap (--cost-cap); a free estimate (--estimate-only) is available.